lxyzler

> @lxyzler 您好，请问您是只用强监督训练么？(是不是意味着没有使用弱标签部分) 是的

> @lxyzler 请问您训练这个模型的效果怎么样？为什么我训练了几天的模型效果几乎没有，mask图像很糟糕 我训练了三个星期，在icdar15上的准确率大概百分之四十几，效果不是很理想

Linux xiaoi2018-754 3.10.0-957.1.3.el7.x86_64 #1 SMP Thu Nov 29 14:49:43 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux +-----------------------------------------------------------------------------+ | NVIDIA-SMI 410.78 Driver Version: 410.78 CUDA Version: 10.0 | |-------------------------------+----------------------+----------------------+ | GPU Name...

I have rewrite the data_generator File "/home/sherry/CRNN-master/utils.py", line 162, in evaluate ctc_decode = K.ctc_decode(y_pred[:, 2:, :], input_length=np.ones(shape[0])*shape[1])[0][0] I try the Synth90k,in this case,y_pred[:, 2:, :].shape is (5000, 48, 37) if...

@karolmajek @mcahny I also want to run the demo on new video without GT results. Do you have any code to share?

> @lxyzler > > can you please check this [link](https://stackoverflow.com/questions/56528061/deeplab-predict-a-black-mask) and see if it helps you. > Request you to share colab link or code snippet to reproduce the issue...

@ravikyram @jvishnuvardhan Do you have any ideas, or what should I pay attention to while training without loading the pre training model